Intelligent Controller for Containerized Applications

1. A system comprising: a plurality of storage devices configured to store data associated with at least one of homogeneous and heterogeneous applications running in containers; and a physical component comprising a programmed controller configured to balance workloads of the containers by: grouping the containers based on characteristics of the workloads of the containers, and grouping active and idle containers to one of the storage devices based on time to live (TTL) of the active and the idle containers.

2. The system of claim 1 , further comprising a physical component comprising a programmed detector configured to detect characteristics of a new workload to be assigned to the containers.

3. The system of claim 2 , wherein the physical component comprising the programmed detector is further configured to obtain characteristics of an existing workload running in the containers.

4. The system of claim 3 , wherein the physical component comprising the programmed detector is further configured to assign the new workload to the containers based on the characteristics of the new workload and the existing workload.

5. The system of claim 2 , wherein the physical component comprising the programmed detector is integrated with the physical component comprising the programmed controller.

6. The system of claim 2 , wherein the physical component comprising the programmed controller is further configured to obtain information regarding the assignment of the new workload from the physical component comprising the programmed detector.

7. The system of claim 1 , wherein the physical component comprising the programmed controller is configured to balance the workloads by distributing the containers to one or more of the storage devices.

8. The system of claim 1 , wherein the physical component comprising the programmed controller is configured to balance the workloads to improve quality of service (QoS) in terms of application throughput and latency.

9. The system of claim 1 , wherein the physical component comprising the programmed controller is configured to balance the workloads to improve distribution of the containers, thereby improving utilization of the storage devices.

10. The system of claim 1 , wherein the physical component comprising the programmed controller is further configured to balance storage resource requirements for all containerized applications to ensure fair resource management.

11. The system of claim 1 , wherein the physical component comprising the programmed controller is configured to balance the workloads by grouping all sequential accesses to one of the storage devices or a group of the storage devices.

12. The system of claim 1 , wherein the physical component comprising the programmed controller is configured to balance the workloads by grouping write-intensive and read-intensive containers to one of the storage devices or a group of the storage devices.

13. The system of claim 1 , wherein the physical component comprising the programmed controller is configured to balance the workloads by grouping the containers of same TTL (time to live) to a first storage device of the plurality of storage devices, the first storage device comprising at least two storage devices of the plurality of storage devices.

14. A method of balancing workloads in containers, the method comprising: receiving a new workload; receiving or detecting characteristics of the new workload; obtaining characteristics of existing workloads in the containers; assigning the new workload to the containers based on the characteristics of the new workload and the characteristics of the existing workloads; and balancing the new workload and the existing workloads by grouping active and idle containers to a storage device based on time to live (TTL) of the active and the idle containers.

15. The method of claim 14 , the method further comprising: reading a grouping algorithm; obtaining information on the new workload and the existing workloads of the containers; and assigning the containers to different device or devices based on the information on the new workload and the existing workloads of the containers.

16. The method of claim 15 , wherein the new workload and the existing workloads are balanced by grouping all sequential accesses to a storage device or a group of storage devices.

17. The method of claim 15 , wherein the new workload and the existing workloads are balanced by grouping write-intensive and read-intensive containers to a storage device or a group of storage devices.

18. The method of claim 15 , wherein the new workload and the existing workloads are balanced by grouping the containers of same TTL (time to live) to a storage device.

19. A controller system comprising: a processor, and a memory device, wherein the memory device stores computer executable instructions that, when executed by the processor, cause the processor to balance workload of containerized applications by assigning containers to different device or devices based on information on workloads of the containers and by grouping active and idle containers to one of the devices based on time to live (TTL) of the active and the idle containers.